Proton Pump Inhibitors and Odds of Cholangiocarcinoma

A Retrospective Case-control Study

Jianping Xiong; Yaqin Wang; Weiyu Xu; Zhisong Liu; Haochen Wang; Zhiyuan Zhang; Yanjing Han; Chao Yin; Shasha Cao; Zheran Yang; Tianhao Su; Jian Wei; Guang Chen; Long Jin


Liver International. 2020;40(11):2848-2857. 

In This Article


Patient Characteristics

Basic information for both CCA patients and controls is presented in Table 1. During the research, 1468 CCA patients, together with 2936 controls, were recruited into this study. In the case group, 826 had ICC (56.3%) and 642 had ECC (43.7%). In addition, 463 subjects in the ECC group (72.1%) harboured Klatskin's tumour. There were 834 (56.8%) females in the case group and 1665 (56.7) in the control group, with mean ages of 51.9 and 51.7 years respectively. Compared with the control group, the case group had similar distributions of age, gender, drinking, smoking, cirrhosis, choledocholithiasis or hepatolithiasis, HBV, HCV infection, parasitic infections (hepatobiliary flukes, opisthorchis viverrini and clonorchis sinensis), H pylori infection, cholecystolithiasis, fatty liver disease, hypertension, dyslipidemia, diabetes mellitus, obesity, coronary artery disease, peptic ulcer disease, GERD, metformin use, statin use and aspirin use.

Association Between Proton Pump Inhibitors and CCA Odds

Table 2 indicates the odds of PPI exposure in CCA. In this case-control study, 135 of 1468 (9.2%) CCA patients and 173 of 2936 (5.9%) controls were exposed to PPIs for at least 28 cDDD. When comparing users of PPIs with non-PPI users, we found that PPI use was associated with a 1.61-fold elevated CCA odds (P < .001) (AOR = 1.61, 95% CI = 1.28–2.04; P < .001). In addition, there was no evidence of multicollinearity among covariates, such as peptic ulcer, GERD and H. pylori infection. Maximum VIF (variance inflation factor) was lower than 2.0, and lowest eigenvalue was 0.35. Next, the effects of the PPI dosage and duration on the CCA odds were examined. As a result, for the 28–90, 91–180, and 180 + cDDD group, the AORs were 1.50 (95% CI = 1.02–2.24), 1.65 (95% CI = 1.16–2.17) and 2.07 (95% CI = 1.13–3.08), respectively, relative to the ≤ 27 cDDD group (Table 3). According to the PPI application in the cDDD subgroup, the PPI exposure risk increased significantly in the 180 + cDDD group, demonstrating the highest effect of dose response (P for trend < .001). When stratified by a PPI use duration ≤ 3 or > 3 years, the AORs were 1.58 (95%CI = 1.23–2.01) and 2.23 (1.04–4.76) respectively. In the stratification by CCA subtypes, the AORs of PPIs were consistent for both CCA subtypes, with ORs of 1.36 (AOR = 1.36, 95% CI = 1.02–1.83; P = .003) and 1.95 (AOR = 1.95, 95% CI = 1.46–2.62; P < .001) for ICC and ECC respectively (Table 4). For individual PPIs, omeprazole was the drug with the highest use frequency up to the index date (Table 5). Patients using esomeprazole had the highest CCA odds (AOR = 1.98, 95% CI = 1.16–2.71, P = .015), followed by those using lansoprazole (AOR = 1.82, 95% CI = 1.12–2.33, P = .019), omeprazole (AOR = 1.67, 95% CI = 1.15–2.01, P = .002), pantoprazole (AOR = 1.43, 95% CI = 1.06–2.37, P = .020), together with rabeprazole (AOR = 1.39, 95% CI = 1.02–3.89, P = .017) (Table 5).

Sensitivity Analysis

The other matched case-control cohort that was not exposed to PPI within the past two years (in other words, the extended washout period) was selected prior to study participation for the sensitivity analysis, which provided similar results to the initial data (AOR = 1.35, 95% CI = 1.05–1.76, P = .019) (Table 6). In addition, we also used lung cancer as a negative cancer case and conducted another case-control study to confirm that PPI use was not associated with an increased odds of lung cancer when compared to the matched control patients (AOR = 1.22, 95% CI = 0.86–1.45, P = .340) (Table 6). We further added gastric cancer and hepatocellular carcinoma as an outcome control to support the validity of our findings. As reported in previous works,[30,32] the results of our study indicated that PPI use is not associated with the odds of developing hepatocellular carcinoma (AOR = 1.19, 95% CI = 0.88–1.41, P = .336), but it is associated with increased odds of gastric cancer (AOR = 2.18, 95% CI = 1.58–2.49, P < .001) (Table 6). Another analysis was performed by stratifying patients into two groups: (I) patients with and (II) and without H pylori infection. We found that both groups had significantly higher odds of having CCA than their matched control group (Table 7).